Re: [PATCH v2 1/1] mm/shmem: don't use the memcg of an uncharged folio

From: Qi Zheng

Date: Sat Nov 21 2026 - 05:50:23 EST




On 11/21/26 3:50 AM, Chris Mason wrote:
The shrinker for unused huge tmpfs folios can dereference a NULL memcg
if it races with the migration of the folio at the end of a file.

folio_memcg_list_lru_alloc() assumes that folio_memcg() is never NULL.
But it is NULL until the folio is charged, and a migration target can be
in the page cache before charging.

shmem_unused_huge_shrink() looks the folio up without the folio lock,
so we can race with migration:

__migrate_folio() shmem_unused_huge_shrink()
================= ==========================
__folio_migrate_mapping()
xas_store(newfolio)
filemap_get_entry()
returns newfolio
folio_trylock() fails
shmem_unused_huge_requeue()
folio_memcg_list_lru_alloc()
folio_memcg() is NULL
memcg->kmemcg_id
folio_migrate_flags()
mem_cgroup_migrate()
commit_charge(newfolio)

Fix it by keeping the original objcg around while the inode is isolated.

Fixes: 743e7d025d2f ("mm: shmem: make unused huge shrinker memcg aware")
Suggested-by: Qi Zheng <qi.zheng@xxxxxxxxx>
Signed-off-by: Chris Mason <mason@xxxxxxxxxx>
---

Notes:
v1->v2: adopt Qi Zheng's suggestion to return early in the
failure path

mm/shmem.c | 39 +++++++++++++++++++++++++++++++--------
1 file changed, 31 insertions(+), 8 deletions(-)


LGTM.

Reviewed-by: Qi Zheng <qi.zheng@xxxxxxxxx>

Thanks!